Because of their structures, vertical cavity surface emitting laser devices are characterized by ease of lowering the threshold current and the power consumption. In recent years, oxide-confined vertical cavity surface emitting laser devices have been intensively studied, which devices allow lowering the threshold current and provide a higher-speed response compared to ion-implanted vertical cavity surface emitting laser devices that were previously studied (see Non-patent Document 1, for example).
Oxide-confined vertical cavity surface emitting laser devices have an advantage of having favorable transverse mode confinement provided by an oxide, which results in a stable oscillation mode; however, since the optical confinement by the oxide is too strong, it is difficult to obtain a single fundamental transverse-mode oscillation. Note that an oxide-confined vertical cavity surface emitting laser device is referred to simply as “vertical cavity surface emitting laser device” below.
A widely adopted conventional technique for achieving a single fundamental transverse-mode operation is to provide a small area of an unoxidized region, which is a current injection region (current passage region), so that higher-order transverse modes are confined and do not oscillate. In other words, the technique is to cut off higher-order transverse modes.
Another proposed method for achieving a single fundamental transverse mode operation is to reduce the strength of the transverse mode confinement provided by an oxide. If the strength of the transverse mode confinement is reduced, higher-order mode oscillation is suppressed. In this case, there is no need to make the area of the unoxidized region small, and therefore, both thermal and electrical characteristics are improved. This results in an increase in the saturation power and also an increase in the modulation rate. In order to reduce the strength of the optical confinement by an oxide, conventionally, the oxide is provided at a position away from the active layer, or the oxide is made to be thin.
Vertical cavity surface emitting laser devices may be readily arranged in two-dimensions at high density since each laser device emits laser light orthogonally in relation to its substrate. Accordingly, their applications to high-speed and high-definition electrophotographic systems and the like have begun to be explored. For example, Non-patent Document 2 discloses a printer using a 780-nm band VCSEL array (vertical cavity surface emitting laser array). Patent Document 1 discloses a multi-spot image forming apparatus having a multi-spot light source. In general, higher-speed optical writing can be achieved by using a vertical cavity surface emitting laser device capable of performing high-power operations in a single fundamental transverse mode.
Such a vertical cavity surface emitting laser device includes a current confinement structure in order to increase the efficiency of current influx. A commonly used current confinement structure is formed through selective oxidation of an AlAs (aluminum arsenide) layer (the current confinement structure is also referred to as “oxide current confinement structure” below) (see Patent Document 2, for example). An oxide current confinement structure is obtained by forming, in a precursor structure, a mesa of a predetermined size, in which a p-AlAs layer to be selectively oxidized is exposed along the lateral sides and placing the precursor structure in a high-temperature water vapor atmosphere so that Al is selectively oxidized from the lateral sides in such a manner that a central portion of the mesa remains unoxidized. The unoxidized portion functions as a passage region (current injection region) of the current for driving the vertical cavity surface emitting laser. In this way, current confinement is readily obtained.
Regarding a vertical cavity surface emitting laser, if heat generated in the active layer is rapidly released, a rise in the junction temperature (temperature of the active layer) can be suppressed and a decrease in gain can be prevented. This leads not only to a high output but also to favorable temperature characteristics, and hence longer operating life.
Semiconductor multilayer reflectors are in general made of AlGaAs materials. The thermal conductivity of an AlGaAs material largely varies depending on the Al component, and AlAs has the highest thermal conductivity (see FIG. 65).
Given this factor, it has been proposed that each AlAs low refractive index layer, which is included in a semiconductor multilayer reflector disposed on the heat release path side and is adjacent to the resonator structure, is designed to have an optical thickness larger than usual (see Patent Documents 3 to 5, for example).    [Patent Document 1] Japanese Laid-open Patent Application Publication No. H11-48520    [Patent Document 2] U.S. Pat. No. 5,493,577    [Patent Document 3] Japanese Laid-open Patent Application Publication No. 2005-354061    [Patent Document 4] Japanese Laid-open Patent Application Publication No. 2007-299897    [Patent Document 5] U.S. Pat. No. 6,720,585    [Non-patent Document 1] K. D. Choquette, R. P. Schneider, Jr., K. L. Lear & K. M. Geib, “Low threshold voltage vertical-cavity lasers fabricated by selective oxidation”, Electronics Letters, No. 24, Vol. 30, 1994, pp. 2043-2044    [Non-patent Document 2]H. Nakayama, T. Nakamura, M. Funada, Y. Ohashi & M. Kato, “780 nm VCSELs for Home Networks and Printers”, Electronic Components and Technology Conference Proceedings, 54th, Vol. 2, June, 2004, pp. 1371-1375
In electrophotography or the like, a significant influence is exerted on image quality by the rising behavior of the optical output response waveform of the light source, obtained when a drive current is applied to the light source. The optical output response waveform represents the time change in the optical output, and is hereinafter also referred to as “optical waveform”. For example, image quality may be degraded by a fractional change in light intensity not only during the rise time of the optical waveform but also after the optical output has reached constant light intensity at the beginning of the rise.
This is because parts of an image formed during the rise and fall times of the optical waveform are the contour of the image. If the light intensity changes especially during the rise time of the optical waveform and during a certain time period after the optical waveform can be regarded to have substantially risen, the contour of the image becomes blurred, resulting in poor image quality with lack of visual sharpness.
For example, in the case where 300 μs is required to scan one line on an A4 sheet having a width (lengthwise direction) of about 300 mm, the scan distance in 1 μs is about 1 mm. It is said that the human eye has the highest visual sensitivity for change in image density when the width is 1 to 2 mm. Therefore, if the image density changes over about 1 mm in width, the density change is sufficient to be detected by the human eye, giving an impression of a blurred contour.
Another problem that the present invention addresses relates to the optical thickness of the low refractive index layers in the semiconductor multilayer reflector. If the optical thickness of each low refractive index layer is changed from λ/4 (λ is the oscillation wavelength) to 3λ/4, the absorption of light (hereinafter, also referred to simply as “absorption” for convenience) is increased by three-fold). Within the semiconductor multilayer reflector, the closer to the resonator structure, the stronger the electric field intensity, and therefore, a significant influence of the absorption is exerted. As a result, the methods disclosed in Patent Documents 3 to 5 leave the problem of causing a decrease in the slope efficiency and an increase in the threshold current.